Network Optimization Models

Many critical infrastructures can be represented by a network of interconnected nodes and links. Mathematically sound non-linear optimization techniques can then be applied to these networks to understand their behavior under normal and disrupted situations.  Network optimization models are particularly useful for evaluating transportation system disruption effects on system capacity and the effectiveness of measures to reduce those impacts.

Railroad Network Analysis System (R-NAS)

Using a detailed layout of the primary rail tracks, yards, bridges, etc. in the continental U.S. coupled with commodity movement data from the Department of Transportation, R-NAS provides a capability of studying and understanding the flow of commodities over the nation’s rail infrastructure. The network flow models predict link flow volumes (by commodity group) over the networks, and the corresponding times and distances that commodities encounter in moving from origin points to destinations.

After disruption of a given rail asset, the model attempts to find alternate routes for the delivery of commodities. Delivery time constraints can be placed by the user to determine acceptable delays in delivery times, and the model can provide breakdowns of the types of commodities that do not move given the specific disruption in a scenario.

The model has been used to examine commodity flow disruptions due to destruction of railroad assets, and it has also been used to study policy options concerning the movement of toxic chemicals by rail.

Air Transport Optimization Model (ATOM)

The Air Transport Optimization Model (ATOM) is a network optimization model designed to examine the consequences of a partial or complete outage at a major airport or set of airports for an extended period of time (greater than one week). The model is not intended to guide detailed routing and scheduling decisions for each aircraft by each airline; but to simulate disruptions to air transportation of goods and people that are beyond the normal routing and scheduling changes that airlines make on a daily basis.  ATOM simulations provide a sense of what reasonably can be expected and what is possible through cooperation between the airlines after an incident of this type.

These insights can aid officials in the creation of an overall framework within which the individual air carriers would then operate.

ATOM can be used to answer questions such as:

• If a certain hub airport were no longer accessible, where would that traffic go instead? How many passengers could no longer be accommodated?
• What might happen if an entire FAA region had to be shut down for security reasons?
• What is the optimal re-routing to minimize the lost capacity?